The macOS check spends time the tests do not need in two places. Every atomic write the tests make flushes to stable storage (fsutil's WriteFileAtomic family and the history-index bootstrap call File.Sync, which Go runs as F_FULLFSYNC on macOS, about 8.8 ms per synced write on this disk), though no test asserts anything a flush makes true; and the race step names ./internal/... in import-path order, so go test starts internal/surface/cli, the slowest package under -race, among the last on the three-core runner and the step waits on it as a tail.

The macOS check spends time the tests do not need in two places. Every atomic write the tests make flushes to stable storage (fsutil's WriteFileAtomic family and the history-index bootstrap call File.Sync, which Go runs as F_FULLFSYNC on macOS, about 8.8 ms per synced write on this disk), though no test asserts anything a flush makes true; and the race step names ./internal/... in import-path order, so go test starts internal/surface/cli, the slowest package under -race, among the last on the three-core runner and the step waits on it as a tail.

Grounds

• pursued: the macOS check job shortens by the flush time of every test write and by the race step's single-package tail; a macOS check on the merged tip no shorter than the last green run (23.7 min, race step 888s) would show it wrong